The Direct Heat Treatment of Cement Mill Dust to Increase Its Water

The Direct Heat Treatment of Cement Mill Dust to Increase Its Water-Soluble Potash Content. Albert R. Merz. Ind. Eng. Chem. , 1918, 10 (2), pp 106–1...
0 downloads 0 Views 569KB Size
106

T H E JOURNAL OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY

THE DIRECT HEAT TREATMENT OF CEMENT MILL DUST TO INCREASE ITS WATER-SOLUBLE POTASH CONTENT

Vol.

IO,

No. 2

bulletin. The work presented here deals with a method for making water-soluble t h e “recombined” potash of cement dust. By ALBERT R. MBRZ The recovery of the potash which escapes from t h e Received November 9, 1917 kilns of cement mills was first made by the Riverside When a potash-containing silicate mixed with lime Portland Cement Co., a t Riverside, Cal., using t h e is ignited a t temperatures above IOOO’, the lime dis- Cottrell process of electrical, precipitation. The ceplaces more or less potash which is volatilized. This ment dust recovered a t this plant was found t o contain fact suggests a possible procedure for the immediate upward of go per cent of its potash in the water-soluble commercial production of potash from potash silicates. form. A sample of precipitator dust secured from this Since, however, commercial grades of any insoluble mill was analyzed by the author and found t o give a total potash silicate in quantity contain comparatively small potash content of 10.7 per cent, and a water-soluble percentages of potash no process for recovering potash potash content of 9.8 per cent. The water-soluble from such material can offer much promise of profitable potash of this particular dust was g z per cent of t h e application unless there is also produced some other total potash present in the dust. It was natural t o product of value in addition t o the potash. Fortu- expect t h a t the dust which would be recovered by a nately, and by strange coincidence, the two manu- similar method a t other plants would have a waterfacturing industries of this country which have the soluble potash content approximating the same perlargest outputs are industries wherein raw materials centage of the total potash present. It was found, which contain potash silicates are heated with lime- however, when the Cottrell process was installed a t stone t o temperatures so high t h a t the potash is more the mill of the Security Cement and Lime Company, or less displaced and volatilized. These two indus- a t Hagerstown, Md., t h a t the dust collected contained tries, vuiz., the blast-furnace and Portland cement in- a considerable portion of its potash content in a form dustries, are already firmly established and for them which was not readily soluble in water. A sample of the problem becomes merely one of the successful dust obtained from this plant was found t o contain recovery of the volatilized potash as a by-product. 11.4 per cent total potash, whereas the water-soluble It may be stated t h a t the chief potential source of potash content was but 6.8 per cent. Instead of the potash in this country is the raw material which is or anticipated go per cent or over, this particular sample may be used in these industries. of dust contained but 60 per cent of its potash in Some potash is contained in all cement materials. water-soluble form I t has been found a t another I n a recent publication by this Bureau’ it was shown plant, the Alpha Portland Cement Company a t Cement h a t in the raw mix as fed into the kiln the potash varies ton, N . Y., where installation of the Cottrell process from 0 . 2 0 t o 1.16per cent in the various cement mills for the recovery of dust has been made, t h a t the waterof the United States and Canada, and t h a t the per- soluble potash in the dust recovered constitutes a concentage of this potash driven from t h e kilns in the differ- siderably lower hroportion of the total potash present A sample of the dust reent plants variesfrom 24.5t o 95.9. From the results of t h a n a t the Security plant. t h e analysis of samples of raw mix and of cement from 1 0 2 covered a t this mill was found b y the author t o contain 7.0 per cent total potash and only 2.9 per cent waterplants i t has been calculated t h a t the potash escaping from the kilns of these plants ranges from 0.35 t o 5.14 soluble potash This sample of dust, therefqre, had lbs. per barrel of cement produced with a n average but 41 per cent of its potash present in the waterfor the plants of.this country of 1.93 lbs. On the basis soluble form. The term “water-soluble potash” as used above of a n average production of go,ooo,ooobbls., t h e total potash escaping from the cement plants of this country, refers t o t h a t potash which is obtained in solution b y a s a t present operated, amounts t o about 87,000 tons the procedure given in the ‘(Methods of the Associaannually. The profitable recovery of this potash is tion of Official Agricultural Chemists,” in accordance not dependent upon successful collection alone, but i t with which I O g. of the sample are boiled with 3 0 0 cc. must be obtained in a form so concentrated as t o be of water for 30 min. and the volume subsequently merchantable and a t a cost sufficiently low t o permit brought t o 5 0 o . c ~ . As i t is customary in the fertilizer of a profit under normal market conditions. Three trade t o make payment for none other t h a n this potash factors determine such profitable recovery and must i t becomes a matter of vital importance t o the cement be considered in any attempt advantageously t o ob- manufacturer who recovers his dust for sale t o t h e tain potash as a by-product in the cement industry: fertilizer trade on the basis of its water-soluble potash liberation of the potash, recombination and collection. content t o ascertain the cause of the failure t o secure A full discussion of the various conditions which the maximum amount of his potash in this form, t o affect the liberation of potash from the raw mix in t h e devise methods, if possible, t o obtain most of the potash kiln and of the methods which have been devised t o in his dust directly as water-soluble potash, t o adopt increase the percentage of potash volatilized is t o be some procedure which will profitably increase t h e found in the publication already mentioned. Similarly water-soluble potash of his dust as a t present obtained t h e methods of collection of t h e potash t h a t escapes or, finally, t o endeavor t o secure from the fertilizer trade from t h e kilns have received ample treatment in this credit for t h a t available potash of his dust which is but slowly water-soluble and for which a t present 1 W. H. Ross, A. R. Merz and C . R. Wagner, U. S. Department of he receives no compensation. Agriculture, Bull. 678.

Feb., 1918

T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y

As stated in a previous publication,l the potassium compounds occurring in cement dust may be divided into three groups: ( I ) those which are readily watersoluble; ( 2 ) those which are slowly soluble; and (3) those which are insoluble. The insoluble potash represents the combinations occurring in the original silicates of the raw mix carried over niechanically in the dust before being subjected t o a temperature sufficiently high t o bring about decomposition. The form of combination which is slowly soluble in water has been attributed t o the recombination of the volatilized potash with the finely divided incandescent particles of siliceous coal ash carried over in the dust2 and i t is claimed8 t o be proportional t o the amount of coal ash present in the gases from the kiln. Where oil or gas is used for fuel this combination of the potash occurs in comparatively small amount but, where coal is used for burning, the extent t o which the potash occurs in this “recombined” form may be considerable. Of the three cement p’ants mentioned above, the Riverside Portland Cement Company uses oil for fuel while the Security Cement and Lime Company and the Alpha Portland Cement Company use coal in their operation. The insoluble potash is assumed t o be t h a t portion of the total potash which remains undissolved after the dust is boiled in a 5 per cent solution of hydrochloric acid. The difference between the total potash and the sum of the watersoluble and insoluble portions is taken as the slowly soluble or recombined potash. It has been shown by Nestell and Anderson4 t h a t continued extraction of cement dust with boiling water for I O hrs. is sufficient t o dissolve practically all this slowly soluble potash. Various methods have been tried t o prevent the recombination of potash and i t is reported by J. J . P ~ r t e r of , ~ the Security Cement and Lime Company, t h a t the use of salt has been found beneficial in this connection. I n a publication by TV. H. Ross5 it was shown t h a t when Eeldspar and lime are digested with water under a steam pressure of I O t o 5 atmospheres about g o per cent of t h e potash in the feldspar passes into solution. I n cement dust as i t escapes from the kilns, the slowly soluble and insoluble potash are already associa.ted with a considerable percentage of free lime and consequently he concluded t h a t the greater part of the constituents might be recovered in water-soluble form by digesting the dust with steam under pressure. Experimental work has shown this t o be the case and the results obtained b y this prooedure will be published separately. Since ignition of a potash-containing silicate in the presence of lime liberates potash and since the cement dust, a s has already been stated, contains a considerable proportion of free lime, i t occurred t o the author t h a t the recombined potash might be rendered watersoluble by simple ignition of the dust. A preliminary W. H. Ross and A. R. Merz, THIS JOURNAL, 9 (1917). 1035. R . J. Nestell and E. Anderson, Ibzd , 9 (19171, 646.8 J . J. Porter, paper presented at the meeting of the-Portland Cement Association, Chicago, September, 1917. 1 2

4 LOC.

Cit.

S T H I S JOURNAL,

9 (1917), 467.

107

experiment was carried out in which a sample of the above-mentioned treater dust from the Security Cement and Lime Company was ignited with an equal weight of calcium carbonate in a J. Lawrence Smith crucible a t about 1000’for a period of 40 minutes. The watersoluble potash was found t o have been increased by this treatment from 6.8 t o 10.4 per cent on the basis of the original sample. I n other words, the watersoluble potash had been increased from 60 per cent t o 91 per cent of the total potash content. The addition of calcium carbonate served, however, t o considerably

x

a

t

I Temperature

“C.

FIG. I-PERCENTAGES OF WATER-SOWBLE POTASH IN

U N T R E A T E D AND

T R E A T E D DUSTS

lower the percentage of water-soluble potash in the resultant product SO t h a t the latter contained 7.1 per cent of water-soluble potash. It is possible t h a t a procedure somewhat similar t o this on a large scale should prove profitable t o the cement manufacturer who a t present receives nothing for his recombined potash, for although he has not appreciably increased the percentage of water-soluble potash in the material he sells, he has increased t h e quantity of his product over 40 per cent.

I08

T H E JOURNAL OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY

Similar experiments were carried out without the addition of calcium carbonate, using t h e same dust a n d also the dust from the Alpha Portland Cement Company already referred to. The results obtained were unexpected-with the former dust there was a decided increase of t h e water-soluble potash of the original sample, whereas with the latter there was a n actual decrease. Table I gives the percentages of total potash in the dusts before treatment and the percentages of water-soluble potash before and after treatment a t about 1000' in a closed crucible all expressed on the basis of the original sample. TABLE I

CEMENTDUST FROM Security Cement Alpha Portland and Lime Co. Cement Company Original Dust Original Dust Per cent Per cent Total potash KzO, .................... 11.4 7.0 Water-solublk potash before treatment. .. 6 . 8 2.9 Water-soluble potash after treatment. 9.4 2.1

...

,From Table I i t may be seen t h a t ignition of the Security dust in this manner has caused a n increase of water-soluble potash from 6 0 t o 8 2 per cent of the total potash in the un-ignited dust, while a n identical procedure with the Alpha dust has brought about an actual decrease of water-soluble potash from 41 t o 30 per cent. The method of ignition which was carried out as in the J. Lawrence Smith method precluded the loss of this potash by volatilization and the only conclusion t o be drawn was t h a t a further '(recombination'' of water-soluble potash takes place in the dust. TABLE11-EFFECTS ON BASISOF ORIGINAL DUST Loss in weight Tottllpotash,KzO Water-solublepotash ON BASISOF PRODUCT Total potash Water-soluble potash

40 MINUTESIN AN OPEN DISH O N SECURITYDUST Untreated Tepperature of Ignition Dust 600' 700 800' 900' 1000° 5.69 6.05 6.40 6.64 7.32 11.40 11.37 11.42 11.42 11.27 11.19 6.79 9.36 9 . 9 5 10.18 10.55 10.62 12.06 9.93

12.15 10.59

12.20 10.88

IO,

No.

2

controlled by use of a thermo-electric pyrometer and a rheostat in the circuit of the electrically heated muffle, and i t was decided t o make ignitions of the two cement dusts through a range of temperatures t o ascertain the influence of differences of the temperature of ignition on the conversion of "recombined" potash into watersoluble potash or on the further recombination of watersoluble potash. As interest was centered on the behavior of the Alpha dust a n initial experiment w a s run with this dust subjected for 40 min. t o a temperature of 600'. This experiment showed t h a t instead of a diminution of water-soluble potash as obtained previously there was a n increase thereof when tlie carbonaceous matter was removed b y combustion. It is questionable t h a t the recombination of potash can be attributed solely t o the siliceous ash of t h e coal in the light of the above results. The writer hesitates t o ascribe this effect t o the carbon, however, because t h e Alpha dust also contains sulfides in some quantity, a fact forcibly presented when the dust was treated with hydrochloric acid. These sulfides no doubt also undergo change when the dust is ignited in a n oxidizing atmosphere and i t may be t h a t they have an influence on the recombination of potash in t h e absence of air. Table I1 contains the results of this series of experiments with the two dusts. The readings of the pyrometer are probably not exact indications of t h e actual temperatures obtaining

O F IGNITION FOR

........................... ..................... .................. ......................... .................. 11.40 6.79

Vol.

12.07 11.30

12.07 11.46

PRECIPITATOR DUST 1100' 8.70 10.99 10.82 12.04 11.85

EXPRESSED IN PERCSNTAGES

Untreated Dust 600' 12.91 7102 6.98 2.93 4.16 7.02 2.93

8.02 4.78

ALPHADUST Temperature of Ignition 700' 800' 900' 1000' 13.21 13.48 14.55 15.52 6.97 7.00 6.48 5.97 4.69 5.06 4.96 5.15 8.03 5.40

8.09 5.85

7.58 5.81

7.07 6.10

llOOo 17.84 5.38 5.01 6.55 6.10

TABLE 111-WATER-SOLUBLEPOTASHIN IGNITED CEMENTDUSTS EXPRESSED IN PERCENTAGES OF TOTAL POTASHCONTENT Untreated Dust 60 On basis of total potash in original dust.. On basis of total potash in ignited dust..

.. ....

600' 82 82

SECURITY DUST Temperature of Ignition 700' 800° 900' 1000' 87 89 93 93 87 89 94 95

A glance a t the dusts in their original condition before ignition would be sufficient t o enable anyone t o distinguish between them. The Security dust is ashgray in color while the Alpha dust is black, evidently containing unburnt carbonaceous matter. A sample of the latter dust was boiled with concentrated hydrochloric acid for 30 minutes, filtered on an alundum plate and a carbon determination was made on the residue. The carbon was found t o form 9.26 per cent of the dust. Assuming the dust lost from the kiln t o be 4 per cent of the raw mix, and 600 lbs. of raw mix as necessary t o yield a barrel of cement, then t h e loss of carbon per barrel of cement produced amounts t o 2.2 lbs. The view t h a t this carbonaceous matter may have an influence on the behavior of the cement dust suggested the ignition of a sample of Alpha dust in a n open dish placed in a muffle so t h a t the caybonaceous matter might be burned up, combustion of this material not occurring t o any great extent in the closed J. Lawrence Smith crucible. The temperature a t which such ignition occurred could be approximately

llOOo 95 98

Untreated Dust 42

..

600" 59 60

ALPHADUSTTemperature of Ignition 700' 800' 900' 1000° 67 72 71 73 67 72 71 86

llOOo 71 93

in the dishes. I n the first place, i t was evident t h a t all parts of the muffle were not a t a uniformly high temperature and, again, the temperatures locally produced by the combustion of the carbon in the Alpha cement dust may have been considerably in excess of the recorded temperatures. I n consequence of this carbon content the manner of distribution of the sample in the dish may also have had an influence on the local temperatures of the. samples. These same conditions, however, would obtain in large scale operations. Table I11 shows t h a t the ignition a t temperatures of 6 0 0 " or over in a n oxidizing atmosphere of treater dust from cement mills results in a liberation of the recombined potash of the dust and t h a t this released potash (at 900') amounts t o 33 per cent of the total potash of the original dust in the one case and t o 29 per cent in the other. I n view of this fact it is considered possible t h a t a procedure similar t o this may find successful application on a manufacturing scale. The use of a kiln using oil or gas as a fuel is t o be considered necessary for such operations, or if coal is used some arrange-

T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY

Feb., 1918

ment whereby heat is supplied without contact of t h e fuel with the dust t o be treated. I n case the dust is t o be sold directly t o the fertilizer trade such a process would have the advantage of yielding a product with its water-soluble potash more concentrated than in the original dust, for not only is the recombined potash liberated by this procedure but the total weight of the material is diminished. This diminution of weight may be attributed t o loss of water, of carbon dioxide, and of any carbonaceous matter in the dust. If, however, t h e preparation of potash salts is aimed at, i t is possible t h a t for dusts behaving like the Alpha dust digestion under steam pressure, as mentioned above, would be the preferable method of treatment since in the latter method there is no volatilization of potash and the yield of water-soluble potash on the basis of the total original potash is consequently greater. This volatilization apparently commences a t about 900’. As stated before, the temperatures of the Alpha dust very likely were considerably above those indicated by the pyrometer and relatively greater volatilization of potash in this dust a t a given temperature may be easily explained on this assumption. To ascertain the effect of time of heating on the two dusts, samples were ignited a t 1000’ for 20, 40 and 60 minutes. The results obtained are shown in Table IV. TABLE IV-THE EFFECT ON WATER-SOLUBLE AND TOTAL POTASH CONTENT O F VARIATIONS I N DURATION OF IGNITION. TFMPERATURE 1000 --SECURITY DUST---- --ALPHA DUST--20 min. 40 min. 60 min. 20 min. 40 min. 60 min. ON BASXSOF ORIGINALDUST Loss in weight . . . . 7 . 0 6 7 . 3 2 7.37 15.15 15.52 15.80 5.97 5.89 Total potash.. . . . . . . . 11.33 11.19 11.06 6.12 Water-soluble ootash.. 10.60 10.62 10.58 5.15 4.89 5.02

. . .. . .

..

O N B A S I S OF PRODUCT

.

Total potash ..... , , , , 12.19 Water-solublepotash . 11.41

12.07 11.46

11.94 11.42

7.21 5.92

7.07 6.10

7.00 5.81

I t is apparent from inspection of this table t h a t for the periods considered ignition of the dust for a longer period t h a n 2 0 minutes has no decided effect on the ratio of water-soluble t o total potash in the material. S U M MA RY

I-Dusts from cement mills using coal as fuel have considerable proportions of their potash content in a form not readily water-soluble. 11--Ignition of such dusts in an oxidizing atmosphere a t temperatures of 600-1100’ converts the “recombined” potash into a readily water-soluble form. 111--For the periods considered ( 2 0 t o 60 minutes) time of ignition apparently has little effect on the watersoluble potash content of the resultant product. DEPARTMENT OF AGRICULTURE SOILS, WASHINGTON, D. C.

BUREAUOF

EFFECT OF COAL ASH ON THE LIBERATION AND NATURE OF CEMENT MILL POTASH By N. S. POTTER, JR., AND R. D. CHEESMAN Received December 6, 1917

I n all articles thus far published relating t o potash as a by-product in the manufacture of cement, no consideration has been given t o the potash content of the coal used in burning and its attendant effects. It is the purpose of this paper t o point out the effect t h e coal ash has upon the liberation of potash in the

I09

kilns and upon the nature of the so-called “treater dust” collected. The potash collected from the kiln stack gases where coal is used for burning appears in practically two forms, water-soluble potash and the insoluble or slowly soluble potash. The insoluble potash has been attributed t o two causes: the potash in the unburned or partly calcined raw material carried over mechanically in the gases and t o a recombination of the volatilized potash with the finely divided ash particles of the coal. R. J. Nestell and E. Anderson in their paper, “The Nature of Cement Mill Potash,”l state t h a t “the most important differences in the potash material from coalfired and oil-fired kilns, as shown by the analyses given, lie in the relative amounts of soluble and insoluble K20, and in the wide variation in potash concentration in the lighter fractions of dust obtained from kilns using these two different forms of fuel. I n the dust from the oil-fired and coal-fired kilns previously referred to, where in the first case the per cent of insoluble KaO was 0.56 and in the second case 4.55 per cent, i t seems safe t o assume t h a t the amount of mechanically carried-over raw material was practically the same, consequently the difference noted in the amount of siliceous material shown cannot be due t o a greater amount of calcined raw mix in one case. The only other source of siliceous material is the ash from the coal used as fuel. Since this ash, coming as i t does from finely powdered coal, must be in a state of extremely fine subdivisions, approaching t h a t of a true fume, i t is reasonable t o suppose t h a t part of this ash would be collected among the lighter portions of dust. Since approximately 9 lbs. of coal ash are introduced into the kiln per barrel of clinker burned, if only one-half this amount is carried out with t h e gases, it would still be sufficient t o effect appreciably the composition of the collected dust, as the amount of dust caught per barrel of clinker produced is only about 2 0 lbs. Consequently it is probable t h a t t h e considerable amount of insoluble potash shown t o be present in the dust from the coal-fired kilns is in reality due t o a combination of the volatilized potash with the finely divided incandescent particles of siliceous coal ash.” Wm. H. Ross, in his paper on “The Extraction of Potash from Silicate Rocks--II,”2 also states t h a t “the slowly soluble combination is explained on the ground t h a t during the burning of the cement part of the volatilized potash undergoes a recombination with t h e silicates in the dust.” The following is taken from an article on “The Recovery of Water-Soluble Potash as a By-product in the Cement Industry:”3 “The insoluble potash represents the combinations occurring in the original silicates of the raw mix which is carried over mechanically in t h e dust before being subjected t o a sufficiently high temperature t o bring about decomposition. The form of combination which is slowly soluble in water is supposed t o be due t o a recombination of the volatilized potash with the silicates in the dust. I n 1

THISJOURNAL, 9 (1917), 646.

Ibid., 9 (1917), 467. 8 Ibid., 9 (1917), 103.5. 1